Alzheimer's disease (AD) is common and highly heritable with many genes and gene variants associated with AD in one or more studies, including APOE ε2/ε3/ε4. However, the genetic backgrounds for normal cognition, mild cognitive impairment (MCI) and AD in terms of changes in cerebrospinal fluid (CSF) levels of Aβ1-42, T-tau, and P-tau181P, have not been clearly delineated. We carried out a genome-wide association study (GWAS) in order to better define the genetic backgrounds to these three states in relation to CSF levels.
Methods
Subjects were participants in the Alzheimer's Disease Neuroimaging Initiative (ADNI). The GWAS dataset consisted of 818 participants (mainly Caucasian) genotyped using the Illumina Human Genome 610 Quad BeadChips. This sample included 410 subjects (119 Normal, 115 MCI and 176 AD) with measurements of CSF Aβ1-42, T-tau, and P-tau181P Levels. We used PLINK to find genetic associations with the three CSF biomarker levels. Association of each of the 498,205 SNPs was tested using additive, dominant, and general association models while considering APOE genotype and age. Finally, an effort was made to better identify relevant biochemical pathways for associated genes using the ALIGATOR software.
Results
We found that there were some associations with APOE genotype although CSF levels were about the same for each subject group; CSF Aβ1-42 levels decreased with APOE gene dose for each subject group. T-tau levels tended to be higher among AD cases than among normal subjects. From adjusted result using APOE genotype and age as covariates, no SNP was associated with CSF levels among AD subjects. CYP19A1 'aromatase' (rs2899472), NCAM2, and multiple SNPs located on chromosome 10 near the ARL5B gene demonstrated the strongest associations with Aβ1-42 in normal subjects. Two genes found to be near the top SNPs, CYP19A1 (rs2899472, p = 1.90 × 10-7) and NCAM2 (rs1022442, p = 2.75 × 10-7) have been reported as genetic factors related to the progression of AD from previous studies. In AD subjects, APOE ε2/ε3 and ε2/ε4 genotypes were associated with elevated T-tau levels and ε4/ε4 genotype was associated with elevated T-tau and P-tau181P levels. Pathway analysis detected several biological pathways implicated in Normal with CSF β-amyloid peptide (Aβ1-42).
Conclusions
Our genome-wide association analysis identified several SNPs as important factors for CSF biomarker. We also provide new evidence for additional candidate genetic risk factors from pathway analysis that can be tested in further studies.
Background
Alzheimer's disease (AD) is the most common cause of dementia and the most prevalent neurodegenerative disorder. An estimated 10 percent of Americans over the age of 65 and half of those over age 85 have AD. More than 4.5 million Americans currently suffer from the disease. In autosomal dominant early-onset Alzheimer's disease (EOAD, age of onset < 60 years), three susceptible genes (APP, PSEN1, and PSEN2) have been identified [1, 2]. Late-onset AD (LOAD) has ~80% heritability, and is strongly associated with apolipoprotein E (APOE) [3]. APOE has three major alleles (ε2/ε3/ε4) that have different effects on the risk of LOAD, with ε4 having between 10 and 30 times of risk of developing AD by 75 years of age [4].
In addition, several genetic studies have identified putative susceptible loci and genetic variants, including sortilin-related receptor (SORL1) [5, 6], death-associated protein kinase 1 (DAPK1) [7], ubiquilin 1 (UBQLN1) [8], adenosine triphosphate-binding cassette transporter 1, subfamily A (ABCA1) [9], and low-density lipoprotein receptor-related protein 6 (LRP6) [10]. Besides these findings, a large meta-analysis from the AlzGene database [11] reported 598 potential AD-susceptibility genes. For the past few years, genome-wide genotyping association studies brought considerable success by reporting new susceptible loci for AD such as Golgi membrane protein 1 (GOLM1) [12–15]. Recently two groups published the two largest LOAD GWAS [16, 17]; Harold et al. [16] reported the association of SNPs in clusterin (CLU) and phosphatidylinositol binding clathrin assembly protein (PICALM), and Lambert et al. [17] reported association of clusterin (CLU) with LOAD and additionally reported a novel association with complement component (3b/4b) receptor 1 (CR1). These new findings have provided valuable insights in the genetics, neuropathologic mechanisms and pathways associated with AD.
The cerebrospinal fluid (CSF) components β-amyloid peptide (Aβ1-42), total tau protein (T-tau) and phosphorylated tau (P-tau181P) are biomarkers for AD and can be used to aid in diagnosis and to predict progression from mild cognitive impairment (MCI) to AD [18, 19]. These biomarkers can potentially be used in future applications to predict the development of MCI in cognitively normal subjects, progression to AD in MCI patients, and to monitor AD progression [20–23]. These biomarkers may also be used to reveal genes that are important in AD pathogenesis. In the present study, we assessed the several putative AD genes associated with CSF biomarkers that were identified from major public GWAS dataset for Alzheimer's disease, the Alzheimer's disease Neuroimaging Initiative (ADNI). This initiative is the most comprehensive effort to identify neuroimaging measures and biomarkers associated with cognitive and functional changes in healthy elderly people and in people who have MCI and AD [24]. The ADNI data is useful for researchers who are searching for genes that contribute to the development of Alzheimer's disease, which currently affects more than 4.5 million people in the United States alone. We also investigated possible associations of CSF biomarkers (Aβ1-42, T-tau, and P-tau181P) with the number of APOE ε4 allele, age and APOE genotype in order to improve the characterization of the CSF biomarkers and genome-wide SNP genotyping data from the ADNI cohort. This is the first genome-wide association study to use these AD-related biomarkers to identify genes critical the pathogenesis of AD.
Methods
Sample
Data used in the preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database http://www.loni.ucla.edu/ADNI. ADNI was launched in 2003 by the National Institute on Aging (NIA), the National Institute of Biomedical Imaging and Bioengineering (NIBIB), the Food and Drug Administration (FDA), private pharmaceutical companies and non-profit organizations, as a $60 million, 5-year public-private partnership. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of mild cognitive impairment (MCI) and early Alzheimer's disease (AD). Determination of sensitive and specific markers of very early AD progression is intended to aid researchers and clinicians to develop new treatments and monitor their effectiveness, as well as lessen the time and cost of clinical trials.
The Principle Investigator of this initiative is Michael W. Weiner, M.D., VA Medical Center and University of California - San Francisco. ADNI is the result of efforts of many co-investigators from a broad range of academic institutions and private corporations, and subjects have been recruited from over 50 sites across the U.S. and Canada. The initial goal of ADNI was to recruit 800 adults, ages 55 to 90, to participate in the research -- approximately 200 cognitively normal older individuals to be followed for 3 years, 400 people with MCI to be followed for 3 years, and 200 people with early AD to be followed for 2 years. For up-to-date information see http://www.adni-info.org.
The GWAS dataset (downloaded from ADNI website in June 2009) consisted of 243 normal, 235 MCI and 340 AD subjects genotyped using the Illumina Human Genome 610 Quad BeadChips. 410 of these subjects (119 Normal, 115 MCI and 176 AD; 247 males and 163 females) have CSF Aβ1-42, T-tau, and P-tau181P levels. Detailed protocols for subject recruitment and biomarker accrual are available at the ADNI website http://www.adni-info.org/. Demographics, CSF biomarkers, and APOE genotype data of the ADNI dataset are summarized in Table 1, Table 2 and Additional file 1.
Table 1
Demographic, clinical and biomarker data for each subject group after removing 20 outliers (n = 390: Normal, MCI and AD)
Normal (n = 109)
Mean
Age (year)
71.7
CSF Aβ1-42 levels (pg/ml)
205.8
CSF P-tau181P levels (pg/ml)
25.4
CSF T-tau levels (pg/ml)
71.8
Last MMSE
29
N
Male
59
Female
50
MCI (n = 109)
Mean
Age (year)
70
CSF Aβ1-42 levels (pg/ml)
170.4
CSF P-tau181P levels (pg/ml)
33.7
CSF T-tau level (pg/ml)
99.5
Last MMSE
26.5
N
Male
72
Female
37
AD (n = 172)
Mean
Age (year)
71
CSF Aβ1-42 levels (pg/ml)
144.7
CSF P-tau181P levels (pg/ml)
40.3
CSF T-tau levels (pg/ml)
115.3
Last MMSE
21.1
N
Male
105
Female
67
Abbreviations: CSF, cereb al spinal fluid; MMSE, Mini Mental State Examination
Table 2
APOE genotype data for each subject group (Normal, MCI and AD)
APOEgenotype data (after removing outliers: n = 390)
APOEgenotype
ε2/ε2
ε2/ε3
ε2/ε4
ε3/ε3
ε3/4
ε4/ε4
Normal (n = 109)
n
0
18
2
66
21
2
MCI (n = 109)
n
0
10
2
47
40
10
AD (n = 172)
n
0
1
3
54
81
33
APOEallele
ε2
ε3
ε4
Normal (n = 109)
n
20
105
25
MCI (n = 109)
n
12
97
52
AD (n = 172)
n
4
136
117
APOEgenotype data before removing outliers: n = 410)
APOEgenotype
ε2/ε2
ε2/ε3
ε2/ε4
ε3/ε3
ε3/ε4
ε4/ε4
Normal (n = 119)
n
0
20
2
71
24
2
MCI (n = 115)
n
0
10
2
52
41
10
AD (n = 176)
n
0
2
3
54
83
34
APOEallele
ε2
ε3
ε4
Normal (n = 119)
n
22
115
28
MCI (n = 115)
n
12
103
53
AD (n = 176)
n
5
139
120
Abbreviations: CSF, cerebral spinal fluid
Genotyping and sample quality control
Quality control for the genotyping data was performed using PLINK http://pngu.mgh.harvard.edu/~purcell/plink/[25] as follows. 498,205 SNPs were retained after excluding SNPs with Minor Allele Frequency (MAF) < 5%, call rate < 98%, or significant in Hardy-Weinberg Equilibrium test (p ≤ 10-6). All samples had genotyping call rate > 95% and were retained. We then examined population stratification by visual inspection using the first two dimensions from principal components analysis, using SmartPCA from EIGENSTRAT http://genepath.med.harvard.edu/~reich/Software.htm[26, 27]. Self-reported ethnicity and racial identities for ADNI subjects were used to highlight samples in the PCA plot and are summarized in Additional file 2. 390 samples were retained after SmartPCA excluded 20 samples as outliers. We computed the top five principal component coordinates using SmartPCA to correct for stratification in association analysis. SmartPCA removed all but one of the Asian samples and retained Black/African Americans (Additional file 3); Visual inspection suggests that the first principal component (PC0), which explains the most variance in the data, separates the Caucasians and non-Caucasians reasonably well (setting threshold PC0 < = 0.01 can exclude all non-Caucasians). Finally, we excluded 52 non-Caucasian samples as outliers; the genomic control variance inflation factor λ was 1.00983, suggesting minimal population admixture in the final sample used for association analysis. We performed association analysis using age and APOE ε4 genotype as covariate and did not incorporate principal components. Quantile-Quantile plots for each of the three test groups with log10-transformed level of three CSF biomarkers (Additional file 4 and Additional file 5) suggested that population stratification having negligible bias on the genetic associations (Additional file 6). Finally, for the whole analysis we performed in the following method section, the study sample of 390 individuals with three CSF biomarkers was used after removing 20 outliers. The level of three CSF biomarkers was log10-transformed.
Effect of APOEε4 copy number on CSF biomarker levels
We performed Kruskal-Wallis test of log10-transformed CSF biomarkers (Aβ1-42, T-tau, and P-tau181P) stratified by the number of APOE ε4 alleles and the APOE genotypes across all 390 samples or within each of the three diagnostic groups.
Association testing of SNPs and CSF analysis
We tested SNP association with the three CSF biomarkers (Aβ1-42, T-tau, and P-tau181P) by PLINK using samples and SNP markers passing QC. CSF biomarker levels were log10-transformed so they become normally distributed. The association analysis used a full linear model comprising three genetic effects: additive effects of allele dosage (ADD), dominance deviation from additivity (DOMDEV) (negative means the allele is recessive), and 2-df joint test of both additive and dominance (GENO_2DF). In addition, we used age (of being recruited by the study) and APOE ε4 genotype (number of APOE ε4 allele; 0, 1, 2) as covariates after removing outliers. To ensure the significance is not due to population stratification, we also incorporated the top five PCA principal components in the linear regression to further control for the population structure, but found the addition has very little effect on the statistical significance.
Gene ontology and E-SNP analysis
We carried out gene ontology analysis of SNP association results using ALIGATOR (Association LIst Go AnnoTatOR) [28], to find gene-ontology terms enriched with significant SNPs. We used p-value cutoff < 10-3 for SNPs, 5000 replicate gene lists and 1000 permutations as parameters to run ALIGATOR. We examined the top associated SNPs and examined nearby SNPs in linkage disequilibrium (LD) that are associated with gene expression from published eQTL studies [29–33].
Results and Discussion
Association of SNPs with CSF biomarkers
We summarize the top SNPs (p-value < 10-6) with and without using covariates (APOE genotype and age) in Table 3. Without using covariates, we found some genes near the top SNPs, including CYP19A1 (rs2899472, p = 1.86 × 10-7) and TOMM40 (rs2075650, p = 3.03 × 10-7) from Aβ1-42 in normal subject. Several genetic studies have identified those genes as putative susceptible loci and genetic variants associated with Alzheimer's disease [34, 35]. However, close examination of nearby SNPs showed rs2899472 and rs2075650 were not supported by nearby SNPs in LD (nearby SNPs in LD are all non-significant). Because APOE genotypes are strongly associated with AD and TOMM40 is physically close to APOE, we focused on SNPs from adjusted results that consider APOE genotype and age. Here, we found 10 SNPs significantly associated with CSF biomarker level of Aβ1-42, 3 SNPs from T-tau, and 2 SNPs from P-tau181P at 10-6 significance level in normal subjects (Figure 1). In addition, we found 1 SNP significantly associated with CSF biomarker level of Aβ1-42, 3 SNPs from T-tau, and 2 SNPs from P-tau181P at 10-6 significance level in MCI subjects (Figure 2). No SNPs were found at 10-6 significance level in AD subjects (Figure 3). For normal subjects, we found genes near the top SNPs, included CYP19A1 (rs2899472, p = 1.90 × 10-7), NCAM2 (rs1022442, p = 2.75 × 10-7) for Aβ1-42 association and UPP2 (rs2074955, p = 2.07 × 10-7) for P-tau181p association. Again, close examination of SNPs in LD with rs2899472 (CYP19A1) did not support rs2899472. The SNP rs1022442 was in close to genome-wide significance, supported by nearby SNPs (Figure 4), and NCAM2 (neural cell adhesion molecule 2) gene was reported as a genetic factor related to the progression of AD in the Japanese population [36]. The Aβ1-42 level grouped by the SNP rs1022442 genotype over all three cohorts (normal, MCI and AD) supports our finding (Figure 5). Boxplots of Aβ1-42 levels in normal subjects stratified by rs1022442 genotype showed significant differences between AA, AB and BB. Previous study indicated an increased risk associated with rs2899472 in AD patients, which was amplified in APOE ε4 carriers in their study [35]. For MCI subjects, we found several genes near the top SNPs, included FLJ21511 (rs2768975; p = 1.96 × 10-7, rs6850199; p = 3.18 × 10-7) by T-tau association and CHN2 (rs121724, p = 1.45 × 10-7), MTUS1 (rs7842088, p = 2.12 × 10-7) by P-tau181p association.
Table 3
Linear regression result for SNPs with CSF biomarkers in Normal, MCI and AD subjects (Top SNPs with p-value < 10-6)
Linear regression result without using covariates
Normal: Aβ1-42
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
15
rs2899472
49303347
A
GENO_2DF
NA
40.21
1.86E-09
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
15
rs2899472
49303347
A
DOMDEV
0.1791
5.859
5.40E-08
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
1
rs11206801
56623274
A
GENO_2DF
NA
33.14
6.37E-08
1
rs4431886
56624855
C
GENO_2DF
NA
33.14
6.37E-08
15
rs2899472
49303347
A
ADD
-0.1407
-5.699
1.11E-07
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
19
rs2075650
50087459
G
GENO_2DF
NA
30.02
3.03E-07
TOMM40, translocase of outer mitochondrial membrane 40 homolog (yeast)
12
rs10784496
64447238
G
GENO_2DF
NA
29.98
3.09E-07
7
rs12534221
130938530
A
GENO_2DF
NA
27.83
9.04E-07
6
rs1727638
72196293
A
GENO_2DF
NA
27.69
9.71E-07
Normal: T-tau
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
12
rs1997111
117872301
T
GENO_2DF
NA
36.66
1.10E-08
12
rs11069178
117852210
C
GENO_2DF
NA
31.23
1.65E-07
2
rs17189298
119561787
A
GENO_2DF
NA
29.51
3.90E-07
8
rs2935776
109699079
C
GENO_2DF
NA
28.81
5.53E-07
8
rs2928826
109683270
T
GENO_2DF
NA
28.73
5.77E-07
2
rs895401
119584178
G
GENO_2DF
NA
27.86
8.90E-07
2
rs893769
119556554
C
ADD
-0.1024
-5.237
8.92E-07
Normal: P-tau181p
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
18
rs1943816
69209602
C
GENO_2DF
NA
32.09
1.08E-07
2
rs2074955
158686040
C
GENO_2DF
NA
32.06
1.09E-07
UPP2, uridine phosphorylase 2
MCI: T-tau
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
22
rs5998432
31075916
T
GENO_2DF
NA
27.91
8.696E-07
MCI: P-tau181p
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
2
rs7558386
227270383
G
GENO_2DF
NA
32.09
1.08E-07
2
rs7558386
227270383
G
DOMDEV
0.242
5.343
5.27E-07
3
rs7631605
37209593
T
GENO_2DF
NA
27.71
9.60E-07
AD: Aβ1-42
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
21
rs239713
27618347
T
GENO_2DF
NA
29.03
4.96E-07
AD: T-tau
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
20
rs4925189
59380439
G
DOMDEV
0.1969
5.178
6.35E-07
CDH4, cadherin 4, type 1, R-cadherin (retinal)
AD: P-tau181p
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
4
rs12643654
96378840
G
GENO_2DF
NA
29.04
4.95E-07
UNC5C, unc-5 homolog C (C. elegans)
Linear regression result using APOE genotype and age as covariates
Normal: Aβ1-42
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
15
rs2899472
49303347
A
GENO_2DF
NA
53.98
1.90E-12
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
15
rs2899472
49303347
A
ADD
-0.1393
-6.925
3.85E- 0
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
15
rs2899472
49303347
A
DOMDEV
0.1614
6.425
4.15E-09
CYP19A1, cytochrome p450, family 19, subfamily a, polypeptide 1
21
rs1022442
21277717
A
GENO_2DF
NA
30.21
2.75E-07
NCAM2, neural cell adhesion molecule 2
10
rs2493624
19102262
G
GENO_2DF
NA
28.61
6.14E-07
10
rs11015839
19042893
A
GENO_2DF
NA
28.51
6.45E-07
ARL5B, adp-ribosylation factor-like 5b
10
rs11015860
19045045
G
GENO_2DF
NA
28.51
6.45E-07
1
rs10911736
183689227
A
GENO_2DF
NA
28.16
7.69E-07
10
rs11015839
19042893
A
ADD
-0.1151
-5.257
7.87E-07
10
rs11015860
19045045
G
ADD
-0.1151
-5.257
7.87E-07
10
rs11015939
19057948
A
ADD
-0.1243
-5.219
9.25E-07
10
rs2495832
19114467
G
ADD
-0.1243
-5.219
9.25E-07
6
rs12203791
67778209
T
GENO_2DF
NA
27.68
9.76E-07
10
rs10763625
19074957
T
ADD
-0.124
-5.206
9.78E-07
Normal: T-tau
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
12
rs1997111
117872301
T
GENO_2DF
NA
30.74
2.11E-07
2
rs17189298
119561787
A
GENO_2DF
NA
27.92
8.66E-07
18
rs1021599
28469054
T
GENO_2DF
NA
27.88
8.81E-07
Normal: P-tau181p
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
18
rs1943816
69209602
C
GENO_2DF
NA
31.43
1.50E-07
2
rs2074955
158686040
C
GENO_2DF
NA
30.78
2.07E-07
UPP2, uridine phosphorylase 2
MCI: Aβ1-42
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
10
rs2986971
30136634
G
GENO_2DF
NA
31.88
1.20E-07
10
rs2986971
30136634
G
ADD
0.08328
5.228
8.88E-07
MCI: T-tau
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
4
rs2768975
48724658
G
GENO_2DF
NA
30.89
1.96E-07
FLJ21511, hypothetical protein flj21511
4
rs6850199
48695607
A
GENO_2DF
NA
29.93
3.18E-07
FLJ21511, hypothetical protein flj21511
18
rs1445093
48024821
C
GENO_2DF
NA
28.14
7.75E-07
MCI: P-tau181p
Chromosome
SNP
Position
A1
TEST
BETA
STAT
P-value
Gene Symbol, Gene Name
7
rs121724
29207970
G
GENO_2DF
NA
31.49
1.45E-07
CHN2, chimerin (chimaerin) 2
8
rs7842088
17629409
G
GENO_2DF
NA
30.73
2.12E-07
MTUS1, mitochondrial tumor suppressor 1
Abbreviations: SNP, single-nucleotide polymorphism; ADD, additive effects of allele dosage; DOMDEV, dominance deviation from additivity, rather specifying that a particular allele is dominant or recessive; GENO_2DF, a 2 df joint test of both additive and dominance; A1, tested allele (minor allele); BETA, regression coefficient; STAT, coefficient t-statistic
Figure 1
Manhattan plots of the quantitative trait (CSF biomarkers: T-tau, P-tau181Pand Aβ1-42) genome wide association analysis in normal subjects. Colors on x-axis indicate an autosomal chromosome (from chromosome 1 to chromosome 22). The y-axis indicates p-values (-log10(observed p-values)). Red arrows indicate rs2899472 on chromosome 15 and blue arrow indicates rs1022442 on chromosome 21.
Figure 2
Manhattan plots of the quantitative trait (CSF biomarkers: T-tau, P-tau181Pand Aβ1-42) genome wide association analysis in MCI subjects. Colors on x-axis indicate an autosomal chromosome (from chromosome 1 to chromosome 22). The y-axis indicates p-values (-log10(observed p-values)).
Figure 3
Manhattan plots of the quantitative trait (CSF biomarkers: T-tau, P-tau181Pand Aβ1-42) genome wide association analysis in AD subjects. Colors on x-axis indicate an autosomal chromosome (from chromosome 1 to chromosome 22). The y-axis indicates p-values (-log10(observed p-values)).
Figure 4
Upper image shows genes in linkage disequilibrium with the SNP rs1022442 and lower plot shows the distribution of SNP rs1022442 and nearby SNPs. In lower plot, the x-axis indicates loci on chromosome 21 and y-axis indicates p-values (-log10(observed GENO_2DF p-values)). Green arrow indicates rs1022442 (p = 2.75 × 10-7) on loci 21277717 and pink arrow indicates rs2826629 (p = 8.40 × 10-5) on loci 21265887.
Figure 5
Boxplots of Aβ1-42levels in normal, MCI and AD subjects stratified by rs1022442 genotype. The x-axis indicates AA, AB and BB respectively.
Association of APOEwith CSF biomarkers
Previous studies suggesting that CSF Aβ1-42 and T-tau levels are correlated with the number of APOE ε4 alleles [37]. We analyzed the distribution of levels stratified by diagnosis and the number of APOE ε4 in the ADNI cohort after QC (Figure 6) and reached the same conclusion. In the AD group, Aβ1-42 level was inversely correlated with APOE ε4 allele dose. The APOE e4 was not associated with T-tau or P-tau181P levels. Analysis of APOE genotypes showed that ε4/ε4 is associated with CSF biomarker level of Aβ1-42 , T-tau and P-tau181P (Figure 7).
Figure 6
Boxplots of theAPOEε4 copy number with CSF biomarkers in normal, MCI and AD subjects. The x-axis indicates number of APOE ε4 alleles and y-axis indicates CSF biomarkers. P-values are produced by the Kruskal-Wallis test.
Figure 7
Boxplots of theAPOEgenotype with CSF biomarkers in normal, MCI and AD subjects. The x-axis indicates five different APOE genotypes and y-axis indicates CSF biomarkers. P-values are produced by the Kruskal-Wallis test.
Pathway analysis of CSF biomarkers
We ran ALIGATOR to identify top gene ontology terms associated with genes containing SNPs with higher statistical significance, and summarized the results in Additional file 7. We found cerebral cortex development, methionine metabolic process, actinin binding, and pallium development to be among the most significant gene ontology terms associated with CSF biomarker level of Aβ1-42 in normal subjects. Elevation in Aβ in the cerebral cortex has been implicated in the pathophysiology of AD but its mechanism of action is unknown [38]. It has been known that mammals have a fully developed cortex, but the structure it evolved from pallium which is present in all vertebrates as well as the most primitive ones [39]. The medial pallium forms the precursor of the hippocampus. Since hippocampal disruption is one of the earliest signs for AD, pallium development might be involved in the pathophysiology AD.
E-SNP analysis of CSF biomarkers
We collected eSNPs (SNPs known to be associated with the expression level of some genes) in several published expression quantitative linkage (eQTL) studies, and examined regions around significant SNPs (p < 10-4) (Additional file 8). We found significant SNPs from Dixon et al. [29], Stranger et al. [31], and Gibbs et al. [33], but we could not find any top SNPs that are associated with gene expression from other papers [30, 32]. Dixon et al. [29] used lymphoblastoid cell lines (LCLs) derived from children both with and without asthma.
NCAM2 (Neural Cell Adhesion Molecule 2)
We obtained strong evidence that NCAM2 genotypes are associated with Aβ1-42 levels. NCAM2 (Ensemble: ENSG00000154654, OMIM: 602040;) is a 541Kb gene at 21q21.1 with no known alternative spicing forms. Novartis SymAtlas human tissue survey shows the gene (Affymetrix probeset ID 205669_at) is ubiquitously expressed and is highly expressed in cardiac myocytes, blood cells, and appendix. Among neuronal tissues NCAM2 has higher expression levels in prefrontal cortex, superior cervical ganglion, and hypothalamus. The transcript (Ensembl: ENST00000400546) consists of 18 exons and encodes a 93 k-Da, 835-residue plasma membrane protein (NP_004531.2). The NCAM2 protein architecture includes 5 IgC2 (Immunoglobulin C-2 type) domains followed by two FN3 (fibronectin type 3) domains and a transmembrane domain. The gene is conserved in chimpanzee, dog, cow, mouse, rat, chicken, zebrafish, and fruit fly; the eight-domain protein architecture is also conserved in all these organisms except for cow which has only four IgC2 domains. Little is known about NCAM2 except that the protein interacts with prion protein [40] and estrogen receptor 1 (ESR1) [41], is involved in neuron adhesion and fasciculation of neurons, and may be involved in AD [36], prion disease, and Down syndrome [42, 43].
TOMM40 (translocase of outer mitochondrial membrane 40 homolog (yeast)) and CYP19A1 (Cytochrome P450, family 19, subfamily A, polypeptide 1)
It is interesting to note that we found no significant association in APOE (rs769451, chr19: 50102751, p = 0.6682 for association with Aβ1-42 level in normal subjects), but a SNP with strong association in the nearby TOMM40 (intronic SNP rs2075650, chr19:50087459, p = 3.03 × 10-7 for association with Aβ1-42 level in normal subjects), when age and the number of APOE e4 alleles were not included in the regression. The TOMM40 gene is related to how easily molecules can get into and out of the surface of the mitochondria, the energy center of cells. This gene is a transporter of proteins across the mitochondrial membrane, and Sortillin-related receptor, which functions to partition amyloid precursor protein away from β-secretase and -secretase [44]. This is consistent with observations that levels are reduced in the brains of patients with Alzheimer's disease and MCI [44–46]. The TOMM40 gene has been reported in numerous studies in the study of AD genetics; for example, Yu et al. [47] reported possibility that loci in the TOMM40 gene may have a less effect on the risk for LOAD in Caucasians [47], and recently Roses et al. [48] found evidence supporting a poly-T polymorphism (rs10524523, chr19:50094889) in TOMM40 affecting the AD age of onset in two independent clinical cohorts. The potential association of TOMM40 and Aβ1-42 may be how the gene affects the risk and onset age of AD and should be further investigated.
The CYP19A1 gene is localized on chromosome 15q21.2 and spans 123 kb. This gene encodes a member of the cytochrome P450 superfamily of enzymes. Cytochrome P450 aromatase is an enzyme that catalyses the conversion of androgens, such as testosterone, to oestrogens, which act as sex steroid hormones but also function during growth and differentiation. There are high levels of expression in both the gonads and the brain [35]. Huang et al. [35] indicated an increased risk associated with SNP rs2899472 in the total number of AD patients, which was amplified in APOE ε4 carriers.
Conclusions
Our analysis of the ADNI genome-wide association study identified several putative loci that are in genetic association with Aβ1-42, T-tau and P-tau181P levels in cerebrospinal fluids. In particular an intronic SNP rs1022442 of gene NCAM2 is close to genome-wide significance in association with Aβ1-42 in normal subjects. Although the gene is poorly characterized in the literature, prior studies have implicated roles of NCAM2 in prion disease, Down syndrome, and AD. Our findings suggest NCAM2 could be part of the pathway on the pathogenesis of senile plaques in human brains with AD.
With only 119 normal subjects and 410 overall, the GWAS dataset is clearly underpowered. The most significant associations were identified using normal subjects since the variances of the CSF biomarker levels are much smaller in MCI and AD subjects due to dementia. Nonetheless, increasing the number of CSF biomarker measurements is challenging especially for normal subjects. An alternative that will substantially increase the sample size is to examine protein levels in blood instead of CSF, given levels of these proteins in blood are informative about AD pathology or prognosis [49].
Our analysis clearly demonstrates that quantitative trait linkage of biomarkers via genome-wide screening can reveal additional insights into the mechanism that connects known genetic factors to the disease [50]. Moving along this path requires further efforts by the research community towards larger sample size and accrual of additional biomarkers in AD and other neurodegenerative disorders.
Declarations
Acknowledgements
M-RH, GDS, and L-SW are supported by the Alzheimer's Disease Genetic Consortium (U01 AG032984-01 and RC2-AG036528-01). Data collection and sharing for this project was funded by the Alzheimer's Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: Abbott, AstraZeneca AB, Bayer Schering Pharma AG, Bristol-Myers Squibb, Eisai Global Clinical Development, Elan Corporation, Genentech, GE Healthcare, GlaxoSmithKline, Innogenetics, Johnson and Johnson, Eli Lilly and Co., Medpace, Inc., Merck and Co., Inc., Novartis AG, Pfizer Inc, F. Hoffman-La Roche, Schering-Plough, Synarc, Inc., and Wyeth, as well as non-profit partners the Alzheimer's Association and Alzheimer's Drug Discovery Foundation, with participation from the U.S. Food and Drug Administration. Private sector contributions to ADNI are facilitated by the Foundation for the National Institutes of Health http://www.fnih.org. The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer's Disease Cooperative Study at the University of California, San Diego. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of California, Los Angeles. This research was also supported by NIH grants P30 AG010129, K01 AG030514, and the Dana Foundation.
Data used in the preparation of this article were obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database http://www.loni.ucla.edu/ADNI. As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. Complete listing of ADNI investigators is available at http://www.loni.ucla.edu/ADNI/Collaboration/ADNI_Manuscript_Citations.pdf.
Authors’ Affiliations
(1)
Genomics and Computational Biology Program, University of Pennsylvania
(2)
Department of Pathology and Laboratory Medicine, University of Pennsylvania
(3)
Institute on Aging, University of Pennsylvania
(4)
Penn Center for Bioinformatics, University of Pennsylvania
References
Raux G, Guyant-Marechal L, Martin C, Bou J, Penet C, Brice A, Hannequin D, Frebourg T, Campion D: Molecular diagnosis of autosomal dominant early onset Alzheimer's disease: an update. J Med Genet. 2005, 42 (10): 793-795. 10.1136/jmg.2005.033456.View ArticlePubMedPubMed CentralGoogle Scholar
Kowalska A, Pruchnik-Wolinska D, Florczak J, Modestowicz R, Szczech J, Kozubski W, Rossa G, Wender M: Genetic study of familial cases of Alzheimer's disease. Acta Biochim Pol. 2004, 51 (1): 245-252.PubMedGoogle Scholar
Gatz M, Reynolds CA, Fratiglioni L, Johansson B, Mortimer JA, Berg S, Fiske A, Pedersen NL: Role of genes and environments for explaining Alzheimer disease. Arch Gen Psychiatry. 2006, 63 (2): 168-174. 10.1001/archpsyc.63.2.168.View ArticlePubMedGoogle Scholar
Sando SB, Melquist S, Cannon A, Hutton ML, Sletvold O, Saltvedt I, White LR, Lydersen S, Aasly JO: APOE epsilon 4 lowers age at onset and is a high risk factor for Alzheimer's disease; a case control study from central Norway. BMC Neurol. 2008, 8: 9-10.1186/1471-2377-8-9.View ArticlePubMedPubMed CentralGoogle Scholar
Rogaeva E, Meng Y, Lee JH, Gu Y, Kawarai T, Zou F, Katayama T, Baldwin CT, Cheng R, Hasegawa H, et al: The neuronal sortilin-related receptor SORL1 is genetically associated with Alzheimer disease. Nat Genet. 2007, 39 (2): 168-177. 10.1038/ng1943.View ArticlePubMedPubMed CentralGoogle Scholar
Lee JH, Cheng R, Schupf N, Manly J, Lantigua R, Stern Y, Rogaeva E, Wakutani Y, Farrer L, St George-Hyslop P, et al: The association between genetic variants in SORL1 and Alzheimer disease in an urban, multiethnic, community-based cohort. Arch Neurol. 2007, 64 (4): 501-506. 10.1001/archneur.64.4.501.View ArticlePubMedPubMed CentralGoogle Scholar
Li Y, Grupe A, Rowland C, Nowotny P, Kauwe JS, Smemo S, Hinrichs A, Tacey K, Toombs TA, Kwok S, et al: DAPK1 variants are associated with Alzheimer's disease and allele-specific expression. Hum Mol Genet. 2006, 15 (17): 2560-2568. 10.1093/hmg/ddl178.View ArticlePubMedGoogle Scholar
Bertram L, Hiltunen M, Parkinson M, Ingelsson M, Lange C, Ramasamy K, Mullin K, Menon R, Sampson AJ, Hsiao MY, et al: Family-based association between Alzheimer's disease and variants in UBQLN1. N Engl J Med. 2005, 352 (9): 884-894. 10.1056/NEJMoa042765.View ArticlePubMedGoogle Scholar
Sundar PD, Feingold E, Minster RL, DeKosky ST, Kamboh MI: Gender-specific association of ATP-binding cassette transporter 1 (ABCA1) polymorphisms with the risk of late-onset Alzheimer's disease. Neurobiol Aging. 2007, 28 (6): 856-862. 10.1016/j.neurobiolaging.2006.04.005.View ArticlePubMedGoogle Scholar
De Ferrari GV, Papassotiropoulos A, Biechele T, Wavrant De-Vrieze F, Avila ME, Major MB, Myers A, Saez K, Henriquez JP, Zhao A, et al: Common genetic variation within the low-density lipoprotein receptor-related protein 6 and late-onset Alzheimer's disease. Proc Natl Acad Sci USA. 2007, 104 (22): 9434-9439. 10.1073/pnas.0603523104.View ArticlePubMedPubMed CentralGoogle Scholar
Bertram L, McQueen MB, Mullin K, Blacker D, Tanzi RE: Systematic meta-analyses of Alzheimer disease genetic association studies: the AlzGene database. Nat Genet. 2007, 39 (1): 17-23. 10.1038/ng1934.View ArticlePubMedGoogle Scholar
Li H, Wetten S, Li L, St Jean PL, Upmanyu R, Surh L, Hosford D, Barnes MR, Briley JD, Borrie M, et al: Candidate single-nucleotide polymorphisms from a genomewide association study of Alzheimer disease. Arch Neurol. 2008, 65 (1): 45-53. 10.1001/archneurol.2007.3.View ArticlePubMedGoogle Scholar
Grupe A, Abraham R, Li Y, Rowland C, Hollingworth P, Morgan A, Jehu L, Segurado R, Stone D, Schadt E, et al: Evidence for novel susceptibility genes for late-onset Alzheimer's disease from a genome-wide association study of putative functional variants. Hum Mol Genet. 2007, 16 (8): 865-873. 10.1093/hmg/ddm031.View ArticlePubMedGoogle Scholar
Coon KD, Myers AJ, Craig DW, Webster JA, Pearson JV, Lince DH, Zismann VL, Beach TG, Leung D, Bryden L, et al: A high-density whole-genome association study reveals that APOE is the major susceptibility gene for sporadic late-onset Alzheimer's disease. J Clin Psychiatry. 2007, 68 (4): 613-618. 10.4088/JCP.v68n0419.View ArticlePubMedGoogle Scholar
Harold D, Abraham R, Hollingworth P, Sims R, Gerrish A, Hamshere ML, Pahwa JS, Moskvina V, Dowzell K, Williams A, et al: Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer's disease. Nature Genetics. 2009, 41 (10): 1088-U1061. 10.1038/ng.440.View ArticlePubMedPubMed CentralGoogle Scholar
Lambert JC, Heath S, Even G, Campion D, Sleegers K, Hiltunen M, Combarros O, Zelenika D, Bullido MJ, Tavernier B, et al: Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer's disease. Nature Genetics. 2009, 41 (10): 1094-U1068. 10.1038/ng.439.View ArticlePubMedGoogle Scholar
Mattsson N, Zetterberg H, Hansson O, Andreasen N, Parnetti L, Jonsson M, Herukka SK, van der Flier WM, Blankenstein MA, Ewers M, et al: CSF biomarkers and incipient Alzheimer disease in patients with mild cognitive impairment. JAMA. 2009, 302 (4): 385-393. 10.1001/jama.2009.1064.View ArticlePubMedGoogle Scholar
Engelborghs S, Sleegers K, Cras P, Brouwers N, Serneels S, De Leenheir E, Martin JJ, Vanmechelen E, Van Broeckhoven C, De Deyn PP: No association of CSF biomarkers with APOEepsilon4, plaque and tangle burden in definite Alzheimer's disease. Brain. 2007, 130 (Pt 9): 2320-2326. 10.1093/brain/awm136.View ArticlePubMedGoogle Scholar
Hansson O, Zetterberg H, Buchlave P, Londos E, Blennow K, Minthon L: Association between CSF biomarkers and incipient Alzheimer's disease in patients with mild cognitive impairment: a follow-up study. (vol 5, pg 228, 2006). Lancet Neurol. 2006, 5 (4): 293-293.View ArticleGoogle Scholar
Andreasen N, Blennow K: CSF biomarkers for mild cognitive impairment and early Alzheimer's disease. Clin Neurol Neurosur. 2005, 107 (3): 165-173. 10.1016/j.clineuro.2004.10.011.View ArticleGoogle Scholar
Zetterberg H, Wahlund LO, Blennow K: Cerebrospinal fluid markers for prediction of Alzheimer's disease. Neurosci Lett. 2003, 352 (1): 67-69. 10.1016/j.neulet.2003.08.011.View ArticlePubMedGoogle Scholar
Wahlund LO, Blennow K: Cerebrospinal fluid biomarkers for disease stage and intensity in cognitively impaired patients. Neurosci Lett. 2003, 339 (2): 99-102. 10.1016/S0304-3940(02)01483-0.View ArticlePubMedGoogle Scholar
Mueller SG, Weiner MW, Thal LJ, Petersen RC, Jack C, Jagust W, Trojanowski JQ, Toga AW, Beckett L: The Alzheimer's disease neuroimaging initiative. Neuroimaging Clin N Am. 2005, 15 (4): 869-877. 10.1016/j.nic.2005.09.008. xi-xiiView ArticlePubMedPubMed CentralGoogle Scholar
Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, de Bakker PI, Daly MJ, et al: PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007, 81 (3): 559-575. 10.1086/519795.View ArticlePubMedPubMed CentralGoogle Scholar
Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D: Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006, 38 (8): 904-909. 10.1038/ng1847.View ArticlePubMedGoogle Scholar
Holmans P, Green EK, Pahwa JS, Ferreira MA, Purcell SM, Sklar P, Owen MJ, O'Donovan MC, Craddock N: Gene ontology analysis of GWA study data sets provides insights into the biology of bipolar disorder. Am J Hum Genet. 2009, 85 (1): 13-24. 10.1016/j.ajhg.2009.05.011.View ArticlePubMedPubMed CentralGoogle Scholar
Dixon AL, Liang L, Moffatt MF, Chen W, Heath S, Wong KC, Taylor J, Burnett E, Gut I, Farrall M, et al: A genome-wide association study of global gene expression. Nat Genet. 2007, 39 (10): 1202-1207. 10.1038/ng2109.View ArticlePubMedGoogle Scholar
Heinzen EL, Ge D, Cronin KD, Maia JM, Shianna KV, Gabriel WN, Welsh-Bohmer KA, Hulette CM, Denny TN, Goldstein DB: Tissue-specific genetic control of splicing: implications for the study of complex traits. PLoS Biol. 2008, 6 (12): e1-10.1371/journal.pbio.1000001.View ArticlePubMedGoogle Scholar
Stranger BE, Nica AC, Forrest MS, Dimas A, Bird CP, Beazley C, Ingle CE, Dunning M, Flicek P, Koller D, et al: Population genomics of human gene expression. Nat Genet. 2007, 39 (10): 1217-1224. 10.1038/ng2142.View ArticlePubMedPubMed CentralGoogle Scholar
Verlaan DJ, Ge B, Grundberg E, Hoberman R, Lam KC, Koka V, Dias J, Gurd S, Martin NW, Mallmin H, et al: Targeted screening of cis-regulatory variation in human haplotypes. Genome Res. 2009, 19 (1): 118-127. 10.1101/gr.084798.108.View ArticlePubMedPubMed CentralGoogle Scholar
Gibbs JR, van der Brug MP, Hernandez DG, Traynor BJ, Nalls MA, Lai SL, Arepalli S, Dillman A, Rafferty IP, Troncoso J, et al: Abundant Quantitative Trait Loci Exist for DNA Methylation and Gene Expression in Human Brain. Plos Genetics. 2010, 6 (5): 10.1371/journal.pgen.1000952.Google Scholar
Sager KL, Wuu J, Leurgans SE, Rees HD, Gearing M, Mufson EJ, Levey AI, Lah JJ: Neuronal LR11/sorLA expression is reduced in mild cognitive impairment. Ann Neurol. 2007, 62 (6): 640-647. 10.1002/ana.21190.View ArticlePubMedPubMed CentralGoogle Scholar
Kimura R, Kamino K, Yamamoto M, Nuripa A, Kida T, Kazui H, Hashimoto R, Tanaka T, Kudo T, Yamagata H, et al: The DYRK1A gene, encoded in chromosome 21 Down syndrome critical region, bridges between beta-amyloid production and tau phosphorylation in Alzheimer disease. Hum Mol Genet. 2007, 16 (1): 15-23. 10.1093/hmg/ddl437.View ArticlePubMedGoogle Scholar
Shaw LM, Vanderstichele H, Knapik-Czajka M, Clark CM, Aisen PS, Petersen RC, Blennow K, Soares H, Simon A, Lewczuk P, et al: Cerebrospinal fluid biomarker signature in Alzheimer's disease neuroimaging initiative subjects. Ann Neurol. 2009, 65 (4): 403-413. 10.1002/ana.21610.View ArticlePubMedPubMed CentralGoogle Scholar
Yankner BA, Duffy LK, Kirschner DA: Neurotrophic and neurotoxic effects of amyloid beta protein: reversal by tachykinin neuropeptides. Science. 1990, 250 (4978): 279-282. 10.1126/science.2218531.View ArticlePubMedGoogle Scholar
Aboitiz F, Morales D, Montiel J: The evolutionary origin of the mammalian isocortex: towards an integrated developmental and functional approach. Behav Brain Sci. 2003, 26 (5): 535-552. discussion 552-585PubMedGoogle Scholar
Schmitt-Ulms G, Hansen K, Liu J, Cowdrey C, Yang J, DeArmond SJ, Cohen FE, Prusiner SB, Baldwin MA: Time-controlled transcardiac perfusion cross-linking for the study of protein interactions in complex tissues. Nat Biotechnol. 2004, 22 (6): 724-731. 10.1038/nbt969.View ArticlePubMedGoogle Scholar
Carroll JS, Liu XS, Brodsky AS, Li W, Meyer CA, Szary AJ, Eeckhoute J, Shao W, Hestermann EV, Geistlinger TR, et al: Chromosome-wide mapping of estrogen receptor binding reveals long-range regulation requiring the forkhead protein FoxA1. Cell. 2005, 122 (1): 33-43. 10.1016/j.cell.2005.05.008.View ArticlePubMedGoogle Scholar
Gratacos M, Costas J, de Cid R, Bayes M, Gonzalez JR, Baca-Garcia E, de Diego Y, Fernandez-Aranda F, Fernandez-Piqueras J, Guitart M, et al: Identification of new putative susceptibility genes for several psychiatric disorders by association analysis of regulatory and non-synonymous SNPs of 306 genes involved in neurotransmission and neurodevelopment. Am J Med Genet B Neuropsychiatr Genet. 2009, 150B (6): 808-816. 10.1002/ajmg.b.30902.View ArticlePubMedGoogle Scholar
Paoloni-Giacobino A, Chen H, Antonarakis SE: Cloning of a novel human neural cell adhesion molecule gene (NCAM2) that maps to chromosome region 21q21 and is potentially involved in Down syndrome. Genomics. 1997, 43 (1): 43-51. 10.1006/geno.1997.4782.View ArticlePubMedGoogle Scholar
Sager KL, Wuu J, Leurgans SE, Rees HD, Gearing M, Mufson EJ, Levey AI, Lah JJ: Neuronal LR11/SorLA expression is reduced in mild cognitive impairment. Annals of Neurology. 2007, 62 (6): 640-647. 10.1002/ana.21190.View ArticlePubMedPubMed CentralGoogle Scholar
Rogaeva E, Meng Y, Lee JH, Gu Y, Kawarai T, Zou F, Katayama T, Baldwin CT, Cheng R, Hasegawa H, et al: The neuronal sortilin-related receptor SORL1 is genetically associated with Alzheimer disease. Nature Genetics. 2007, 39 (2): 168-177. 10.1038/ng1943.View ArticlePubMedPubMed CentralGoogle Scholar
Yu CE, Seltman H, Peskind ER, Galloway N, Zhou PX, Rosenthal E, Wijsman EM, Tsuang DW, Devlin B, Schellenberg GD: Comprehensive analysis of APOE and selected proximate markers for late-onset Alzheimer's disease: Patterns of linkage disequilibrium and disease/marker association. Genomics. 2007, 89 (6): 655-665. 10.1016/j.ygeno.2007.02.002.View ArticlePubMedPubMed CentralGoogle Scholar
Roses AD, Lutz MW, Amrine-Madsen H, Saunders AM, Crenshaw DG, Sundseth SS, Huentelman MJ, Welsh-Bohmer KA, Reiman EM: A TOMM40 variable-length polymorphism predicts the age of late-onset Alzheimer's disease. Pharmacogenomics J. 2010, 10 (5): 375-384. 10.1038/tpj.2009.69.View ArticlePubMedGoogle Scholar
Ray S, Britschgi M, Herbert C, Takeda-Uchimura Y, Boxer A, Blennow K, Friedman LF, Galasko DR, Jutel M, Karydas A, et al: Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007, 13 (11): 1359-1362. 10.1038/nm1653.View ArticlePubMedGoogle Scholar
Potkin SG, Guffanti G, Lakatos A, Turner JA, Kruggel F, Fallon JH, Saykin AJ, Orro A, Lupoli S, Salvi E, et al: Hippocampal atrophy as a quantitative trait in a genome-wide association study identifying novel susceptibility genes for Alzheimer's disease. PLoS One. 2009, 4 (8): e6501-10.1371/journal.pone.0006501.View ArticlePubMedPubMed CentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
